Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ntorga/agent-starter-kit --skill architect-design-treegit clone --depth 1 https://github.com/ntorga/agent-starter-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ntorga/agent-starter-kit/architect-design-tree)<a href="https://agentmods.dev/skills/ntorga/agent-starter-kit/architect-design-tree"><img src="https://agentmods.dev/badge/skills/ntorga/agent-starter-kit/architect-design-tree/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ntorga/agent-starter-kit/architect-design-tree"><img src="https://agentmods.dev/badge/skills/ntorga/agent-starter-kit/architect-design-tree.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00030 | $0.01174 |
| Opus 5 | $0.00015 | $0.00587 |
| Sonnet 5 | $0.00006 | $0.00235 |
| Haiku 4.5 | $0.00003 | $0.00117 |
Grade A, and why
architect-design-tree scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
The grill interviews the user over the design tree. This skill builds and extends that tree. The tree is the grill's working state: every decision the feature must settle, mapped with its dependencies, pruned to the selected path, ranked by impact. The Maestro computes the frontier from this file and works the rounds. You never talk to the user.
Procedure
- Initial — no tree exists. Build the tree from the task prompt.
- Re-tree — the tree exists. The user's answers opened a branch the tree does not have. Extend the tree. Do not rebuild it.
- Read the task block. Note the selected path (beginner, tinkerer, or pro) and any new user answers it carries.
- Orient in the codebase before mapping decisions:
- Read
.context.mdfiles in the affected directories. - Read
docs/FEATURE-MAP.mdif it exists. - For the beginner path, infer the default architecture from the maintenance, lightweight, and safe principles (KISS, single responsibility, safe boundaries —
rules/code/general.md).
- Read
- Identify the decisions the feature must settle. Map them as a tree with three branches:
- Business — what the user wants. Acceptance criteria, scope, constraints, success conditions.
- Architecture — how the system is structured. Directory layout, layer separation, frameworks, reference projects.
- Implementation — how the work is organized. Epics, dependencies, parallelizable groups. Prune branches the path does not explore. Beginner: business only. Tinkerer: business and architecture. Pro: all three.
- Apply the quality filter to every candidate decision: does this decision pivot the destination substantially? If the answer does not change the project's direction, remove the node. Prune with the filter, not the ceiling. The ceiling only bounds how large a branch grows — never pad the tree to hit it.
- Check the ceilings. Business: 30 (beginner, tinkerer), 100 (pro). Architecture: 10 (tinkerer), 100 (pro). Implementation: 100 (pro). If a branch exceeds its ceiling, prune until it fits.
- For each node, write:
- The question in plain terms the user can answer alone.
- Dependencies — the node numbers that must settle first. A node may also depend on a fact from step 7.
- Recommendation — the answer you would choose, with the reason, in one or two sentences.
- Impact — high, medium, or low, by how many downstream nodes the answer unblocks.
- List the facts the tree needs from the codebase or the environment in the
## Factssection. A fact is a question only a lookup can answer — a library capability, an existing endpoint, a config key. Number themf1,f2, and so on. - Write the tree to
.memory/plan/<feature-slug>/tree.mdin the format below. Create the directory if it does not exist. - For re-tree: read
tree.md. Add the new nodes with their fields. Add new facts if needed. Keep every settled node and every settled fact as is. Never renumber.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 84 lines · 30 tokens per session scan A ea7c723fa5ca
architect-design-tree is a skill published in the GitHub repository ntorga/agent-starter-kit (142 stars, last pushed 2d ago), licensed MIT. It adds 30 tokens to every session and 1,174 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-13.
Other skills, from other repositories
agent-framework-py-release
Use when cutting a Python release for the microsoft/agent-framework monorepo. Triggers on "bump py versions", "cut a python release", "prepare release PR for python", "release py packages", "bump python to X.Y.Z", or similar requests to bump Python package versions and prepare a release PR. Handles all four lifecycle…
python-package-management
Guide for managing packages in the Agent Framework Python monorepo, including creating new connector packages, versioning, and the lazy-loading pattern. Use this when adding, modifying, or releasing packages.
foundry-hosted-agent-validation
Step-by-step process for validating a Python Foundry hosted agent sample (under python/samples/04-hosting/foundry-hosted-agents/) end to end — running it locally (native runtime and azd ai agent run) and after deploying it to an Azure AI Foundry project with azd. Use this when asked to validate a hosted agent sample.
verify-samples-tool
How to use the verify-samples tool to run, verify, and manage sample definitions in the Agent Framework repository. Use this when adding, updating, or running sample verification.
build-and-test
How to build and test .NET projects in the Agent Framework repository. Use this when verifying or testing changes.
python-feature-lifecycle
Guidance for package and feature lifecycle in the Agent Framework Python codebase, including stage meanings, feature-stage decorators, feature enums, and how to move APIs from one stage to the next.